A time for arguments

I’ve been thinking about arguments a lot. We argue a fair amount on People vs Algorithms. And I’ve come to see that as a positive, so long as we do so with respect, good nature and a willingness to change our minds. I worry we’re societally retreating from arguments. We need to relearn the art of arguing.

We are making arguments a central part of TRB Audience Summit on Sept 17 in NYC. In most of our main stage sessions, I’m going to push our speakers to make their argument about building sustainable media businesses. I want The Atlantic CEO Nick Thompson to make the argument for what an AI-native journalism business looks like. I’ll ask USA Today Co CEO Mike Reed to make the argument for collective action to push back on Google, even if that means cutting off Google’s search crawler. 

We have a great lineup of speakers and participants. My argument to come is it will be an invaluable day for anyone building media businesses. Tickets for media professionals are $995 and for tech/ecosystem professions $1750. 

Get your ticket to TRB Audience Summit

You might have missed it, but the economics world is in an uproar. The Economist published what these days is inevitably called a “hit piece” against well-known MIT economist Daron Acemoglu, leaning heavily on anonymous quotes and even worse, critiques from economics bloggers and Epstein-tainted Larry Summers. It argues that while his work is impressive, it is also spotty. The point of the piece is buried in paragraph 13: “Mr Acemoglu’s arguments about AI are also pessimistic to the point of losing credibility.”

Acemoglu said it “felt like a hit piece.” He got the requisite sympathetic responses, many with the very modern “I’m so very sorry you’re dealing with this.” Imagine the discomfort for a 58-year-old Nobel laureate. 

The Economist’s article, however flawed in execution, is making an argument. Economists are in the thick of the biggest shift in our economy in memory. Acemoglu has been a leading voice in warning that AI won’t lead to a productivity boom.

We are living in a consequential time with many decisions about the direction of the economy and society that requires arguments. Instead of having them – they’re friction, after all – AI boosters want to muscle through their grand project without making the argument that all this is a net benefit for society. Anthropic CEO Dario Amodei thinks the best argument would be actually curing cancer. Couldn’t hurt.

The backlash against data centers is a case in point. Silicon Valley has gone through its own version of Kubler-Ross in casting blame in all directions — on the CCP, Ro Khanna, the deep state, “activists” and all manner of other boogeymen to explain why their grand project polls as less popular than ICE. The National Republican Senatorial Committee warned that “data centers are the anchor” and “the companies need to fix the toxic brand” of AI.

This is yet another second-order impact the first-principles crowd didn’t see coming.

Silicon Valley has retreated into its own information bubble and rebranded PR as New Media. Travis Kalanick has resurfaced and made the rounds of the New Media circuit, save Bill Gurley’s B2G. He sat down with Silicon Valley’s Charlie Rose, David Senra, for nearly two hours of light massaging in an opulent setting. Senra made a stunning statement during the chat: He doesn’t have any friends who aren’t founders. 

These companies and their boosters haven’t made their arguments to the 99.9% of the population. Instead, they’re preaching to the choir to get investment, tech talent and dopamine hits on X. Instead, they should follow their warrior ethos of being the “man in the arena” by encountering the outside world, even through critical interviews and public arguments.

I give Elon Musk credit for getting into an argument with The Economist editor Zanny Minton Beddoes. Elon seemed surprised by someone who was willing to get into an argument with him by marshaling facts in support of a thesis. Ross Douthat grew on me because he gave his guests like Bryan Johnson space to make their argument, then he would stress test them. This approach often leads to remarkable moments that you simply do not get in the New Media sauna. 

We need to have arguments about privacy. Flock is a new hot button issue, despite its license plate reader cameras being in use for several years. The company has viral campaigns that incite people to destroy the cameras as violations of privacy. It now has 120,000 cameras deployed in 6,000 communities. In up to 50 cases, law enforcement officers have improperly accessed data to stalk or improperly monitor people. Flock CEO Garret Langley turned to All-In for a friendly to point to its use in finding missing children and capturing criminals while pinning the misuse of the tech on human frailty and inadequate bureaucratic controls. 

This is an argument that involves the tradeoffs of public safety and not just abuses but whether we want to live in a mass surveillance environment. Better yet, we should argue about whether it’s even possible to have privacy in an age of AI.

We have retreated from arguments. Instead, we have rhetoric, which Plato gave the side eye because it favored skilled speakers over substantive issues.

There are endless debates of the Jubilee variety. Candace Owens had one with some guy I never heard of who supposedly paid her $300,000 to get worked over by pro. 

Debate has a winner and a loser. Argument is about getting at truth.

We have retreated from the public square to safe spaces, where algorithms reinforce our biases. Publications aren’t blameless either. They’re often too scared of losing subscribers to actually challenge them with arguments that run counter to their identitarian beliefs.  

That critical role of arguments is overlooked. The aftermath of the apparent suicide of Cambridge professor Jason Ardslay after his qualifications, background and work was called into question has predictably led to a round of recrimination against the criticism. His death is obviously a tragedy, of course, but the public sphere has always had the rough and tumble. 

We pretend like we want politics to be above argument. Sometimes, particularly in this information environment, we get to necessary arguments in weird ways. Jon Ossoff had a breakthrough moment by subtly dropping “Natalie” in an attack against Trump, invoking the name of his aide Natalie Harp, nicknamed the Human Printer for carrying around printouts of X posts for Trump. Ossoff is using her to get into an argument about Trump’s priorities not lining up with the American people. That’s a good political argument. 

Incivility is a feature of American politics, not a bug. It can be used for good and bad.

We shrink from arguments at our own peril. One of the legacies of Covid is that authorities tried to shut down arguments about the disease that in retrospect likely needed to happen. AOC might want to rollerskate away from Woke 1 as cray-cray, but Brett Stephens is right we should have arguments about what impact it had to individuals caught up in the frenzy. The Times did James Bennet dirty because it was scared of an argument put forward by an elected U.S. senator. Yes, Woke 1 was crazy. 

Arguments can be exhausting. They can slow down processes. The Chinese aren’t arguing about AI, we’re told. Good luck to them. Open societies operate differently. We need to have space for arguments, even if it seems like the last thing we need is more discord. A good argument is rooted in good faith, shared evidentiary standards and, critically, an openness to changing your mind.

Although AI adoption is widespread, organizational maturity remains more measured. Just under one-third (30%) of respondents describe their organizations as advanced in AI readiness, while the largest share (44%) characterize themselves as moderately prepared, and more than one-fifth (21%) say they are still in the early stages of adoption. 

See more findings from the latest research from The Rebooting and WordPress VIP

Per Sendall, co-head of advertising at Spotify, explains why "the vast majority" of Spotify's code is now being written by software agents, how the company built and open-sourced a CLI plug-in that lets brands buy audio ads using natural language, and why he thinks agentic negotiation is still years away — because AI models are "hard coded to be nice," making them easy for human negotiators to exploit.

Also covered:

See the interview with Per on YouTube or listen to it 

Send me a note with your feedback by hitting reply.